Appeared: 1993 "Identity and Institutions: The Social Construction of Trade Unions in the United States and Germany in the Nineteenth Century" in Studies in American Political Development, fall 1993, 371-394

Identity and Institutions The Social Construction of Trade Unions in Nineteenth Century Germany and the United States1

DRAFT Gary Herrigel Department of Political Science University of Chicago 5828 S. University Chicago, Il, 60637 First Draft: March 9, 1993 Second Draft: June 10. 1993

1

The aim of this paper is to begin to develop the idea that trade unions are historically constructed as much through considerations of social identity as they are through calculations of economic self interest, market power or functional adaptation in the face of changes in the division of labor.

By social identity, I mean the desire for group distinction, dignity and place within historically specific discourses (or frames of understanding) about the character, structure and boundaries of the polity and the economy. Institutions such as trade unions, in other words, are constituted through and by particular understandings of the structure of the social and political worlds of which they are part. In making this argument, it should be immediately said that I in no way intend to claim that trade unions are only to be understood through the lens of identity or that they do not engage in strategic calculation either in labor markets or in the broader political economy. The point is that action along the latter lines presupposes some kind of commitment on, and even resolution of, issues concerning the former. The discussion below focuses on the emergence of trade union movements in the United States and Germany during the nineteenth and early twentieth centuries. It attempts first to develop the two cases as constituting a paradox and then, second, explains the paradox with an argument about identity.

The paradox is the following: How is it that the country more dramatically transformed by mass production and large scale industry than any other in the world before World War One (the

United States) developed a trade union movement dominated by craft unionism, whereas the country whose industrial producers were perhaps more oriented to craft based production than any others during the same period (Germany) developed trade unions organized as industrial unions? Why weren't the union structures reversed in the two cases? My argument will be that 2 this paradox can be explained by the way in which trade union structure was bound up with efforts to resolve identity and social self definition dilemmas among groups of workers during the profound transformation of the social structure that accompanied late nineteenth century industrialization in both the German and American political economies. In both countries, the union movements ultimately understood themselves to be the representatives of the industrial working class, but each union movement conceptually located and categorized industrial workers differently within their respective societies. Differences in trade union organizational forms expressed the different way in which the social identity of the industrial worker was constructed in each society.

Trade Union Structure and Industrialization

This section develops and critiques a view of the relationship between industrialization and organizational form which I, perhaps unfairly, take to be the standard view. It is, in any case, the most systematic way to view the relationship between industrial development and organization without taking into consideration questions of identity. Standard accounts of trade union development in Germany and the United States emphasize structural and material factors, such as calculations of material interest in the labor market or functional adaptation of organization in the face of changes in the division of labor. On this view, craft unions in the United States and industrial unions in Germany before World War I appear paradoxical because industrial workers in the United States were far more likely to be employed in factories engaged in large scale, high volume, standardized production than was true of German industrial workers during the same period. Most attempts to explain this paradox resort either explicitly or implicitly to the idea of a "lag" or to the notion that industrial union formation was blocked. Some argue that craft union strategies in the labor market continued to be rational for skilled workers, while others resort to exogenous factors such as hostile employers or an unaccommodating state to explain the absence of industrial unions. All of these efforts to resolve the paradox, however, do not hold up under comparative scrutiny. After demonstrating this, the rest of the note will then attempt to develop 3 an alternative argument about social identity and trade union formation that better explains the two outcomes.

The Anatomy of Standard Accounts

Most accounts of the development of trade union organization in the United States and Germany situate their explanations within a larger understanding of the historical process of industrialization in which development is driven by the endogenous unfolding of the division of labor. Within such an understanding of industrialization (of which there are both liberal and marxian variants) the division of labor both determines the appropriate organizational forms that will exist in the economy and undermines those forms that cease to be useful at a new stage of evolution. For example, small firms and craft production are on this view understood to be appropriate to early forms of "free market" based capitalism. In "later" stages the conceptual framing of small firms changes: they become backward, secondary or competitively inferior to more efficient large-scale forms of production characteristic of "modern" or "monopoly" or

"organized" or "corporate" capitalism.2 In a similar way, craft unionism is coded as distinctive of early competitive and artisinal forms of capitalism, whereas industrial unionism is thought to correspond to modern large scale capitalist production.3

The key mechanism of change in such arguments is usually one of three non-political variables which seem to exist behind or underneath the organizational forms in question: depending on the view, the extension of markets, the development of large scale technologies, or the balance of class forces.4 Organizations in society, such as trade unions or industrial corporations (or states), "adapt" or "adjust" to pressures that change in the underlying division of labor generate.

Many newer works in this tradition go to great lengths to emphasize that adjusting organizations have some autonomy in the way in which they restructure themselves in the context of these deeper historical imperatives in the economy. This makes it possible to account for considerable organizational variety cross nationally, despite comparable levels of economic development. 4

Such newer views, however, never really quite reject the idea that there are deeper historical imperatives driving the development of the economy to which adjustment is called for.5

Within this broad conceptual frame, the evolution of trade union organizational form in the

United States is narrated as a slow, uneven and arduous shift toward industrial unionism. Craft unions emerge, naturally, within the transformation of the artisan economy from a pre-capitalist economy in which relations were governed by mutual understandings of honor and brotherhood into a capitalist one in which entrepreneurial masters increasingly separated themselves from wage labor dependent journeymen and apprentices.6 As the artisan economy was undermined by the emergence first of the factory system and then, second, by the large scale production processes of fin de siecle industrial corporations, craft unions struggled to retain their control over production in the new environment through the development of extensive work rules and jurisdictional boundaries.7

The relentless diffusion of large scale factory work, the combination of different trades within single plants and the explosive growth of unskilled factory labor unavoidably embroiled the AFL unions in myriad jurisdictional disputes, and complicated amalgamations.8 Moreover, even though the AFL itself, at least by the beginning of the twentieth century, recognized the value in organizing the unskilled, the fundamental principle of voluntarism that structured federation politics hampered its ability to encourage such organizing: In this sense, craft egoism was structurally protected within the AFL. All of these things slowed down and even blocked the adjustment of the craft union structure to the new situation. Thus it wasn't until the late 1930s and the creation of the CIO unions, decades after the emergence of the large scale industrial corporation, that industrial unionism established a solid foothold in the United States.9

In the standard literature on the German case, the same conceptual practice of matching organization forms to the division of labor also produces industrial unions, but in a very different 5 way: there were no lags or blockages to hinder development as in the American case. The

German arguments begin with the claim that because guild structures in the German artisinal economy were so strong, craft unions were slow to develop in early German capitalism.10

Indeed, not until the pace of industrial growth began to quicken after 1848 did independent unions begin to form at all, and then they arose not out of journeymen's associations as in the US, but out of friendly societies, workers clubs and the solidarity that spontaneous strikes created.11

The new unions were craft unions and continued to be deeply rooted in an artisinal milieu: despite their independence from older guild organizations, Juergen Kocka notes that "to many workers self-organization and collective self-help appeared to be a quasi-natural way of protection against the insecurities of the market economy and the superiority of the employers."12

During the last half of the 19th century, two things occurred which paved the way to the formation of German industrial unions: Industry and large scale factory production spread rapidly throughout the economy, either undermining or pushing smaller craft producers to the margins.13 At the same time, the German state in 1878 initiated a twelve year long campaign of repression against all socialist organizations in the society, including many of the newly formed craft unions. Thus in 1890, when the anti-socialist law was rescinded and trade union organizing once again became possible, the logic of establishing industrial rather than craft unions was extremely compelling, even overdetermined: the requisite preconditions in the division of labor had been established, while state hostility had galvanized solidarity within the working class.14

Although at first only the Metal- and Woodworkers unions established themselves as industrial unions, tremendous industrial growth and concentration, and the emergence of strong employers associations hostile to the unions in the years leading up to the first World War helped to diffuse this organizational form throughout the rest of the union movement. Of Forty-six national unions belonging to the national federation (the Generalkommission) in 1914, seven industrial unions

(the Metalworkers, construction workers, transport workers, factory workers, woodworkers, 6 textile workers and miners) comprised 70 per cent of total membership.15

As familiar--even folkloric-- as these narratives are, a number of problems, both empirical and conceptual, undermine the claim that German and American trade union organization corresponds to the historically regnant structure of the social division of labor. The more the two cases are examined, and particularly when they are compared with one another, the differences in trade union development stop being merely paradoxical and become inexplicable.

Start first with the US case. One difficulty in the standard account is that two important attempts to establish industrial unions (or at least multicraft and multiskill organizations) were made prior to the formation of the CIO: The Knights of Labor and the IWW. Both of these movements were unsuccessful, the latter due to a large extent to state repression, but the former simply fell apart, for reasons that are still highly disputed. Why didn't the more inclusive model advocated by the Knights diffuse and supplant the craft union? The Knights of Labor's conception of a multi-craft and multi-skill union has traditionally been dismissed in the standard literature as a utopian and impractical aberration. Paradoxically, Selig Perlman and Lloyd Ulman both viewed the idea of combining skilled and unskilled in particular as wrongheaded, the former because skilled workers were unwilling to give up their intrinsically stronger bargaining position, the latter because the nationalization of markets created insuperable organizational obstacles to a strategy of organizing multiple crafts and the unskilled.16 Lately it has been suggested that these arguments, though logically sound, have no factual basis in the way the Knights actually fell apart. On this newer view, the real reason for the failure of the Knights was the hostility of employers associations, supported, either tacitly or explicitly, by the government and the courts.17

None of these arguments, however, hold up well when compared to the German case: There crafts aligned with crafts and skilled with unskilled on the argument that such alliances created a 7 stronger bargaining position within the new concentrated economic structure. Moreover, employers association mobilization against the unions is in the German literature given as a stimulus for the solidification of industrial unionism, while hostile courts and labor law seems to have been no deterrent to the emergence of inter-skill solidarity.18 In other words, the same variables produce exactly opposite outcomes. The persistence of the craft union in the United States seems mysterious in the causal logic of the standard accounts which root themselves in an endogenously driven understanding of the development of the social division of labor.

Closer examination of the German case makes the weakness of the explanations for the adoption of union organizational form within this frame even clearer. The most interesting fact above all concerning the early industrial unions in pre World War I Germany was that they organized workers who were overwhelmingly employed in specialized, craft-production oriented, small and medium sized firms, located by and large outside the areas of highly concentrated, large scale

German industry.19 Indeed, in those areas where the processes of centralization and concentration offered to explain the adoption of industrial unionism actually did occur, such as in the Ruhr Valley, or in the large machinery, chemical and electrical firms in Berlin, Nuremberg,

Augsburg, Hannover and elsewhere, the unions were either completely shut out or only weakly represented.20 In the face of this kind of evidence, it is difficult to sustain the claim that the creation of industrial unions in Germany was a rational adjustment to the practical imperatives of an endogenously unfolding process of industrialization.

Trade Union Structure, Industrialization and Social Identity

The alternative approach to the development of trade union form to be presented here avoids the above problems by adopting the approach of a different theoretical tradition, which I will call social constructionism. This tradition completely reverses the lines of causality that structured explanations in what I have called the standard accounts. Thus, rather than view industrialization as an endogenously driven process of the unfolding of the division of labor, social constructionist 8 accounts maintain that the process of industrialization and the development of the division of labor is constituted and shaped exogenously (as it were) by political and social ideas, institutions and relations. In this way organizational construction and reform is seen not as a problem of adaptation or adjustment to processes somehow independent of themselves. Rather organizations in a political economy (labor organizations, corporations, associations, public agencies, etc) are understood to be collectively engaged in the definition and conceptual representation of what those pressures are. For social constructionists, the outcome of struggles among organizations are not conditioned or determined by the social division of labor. Rather, such struggles establish a framework for action and understanding which constitutes what that division of labor is.21

In this framing, industrialization is thus recast as a historically open process. The path of development from past to present is scarred by alternative organizational, technological and political pathways not taken. The stages of the industrialization process (or at least the chronicle of organizational changes in the political economy that we argue over the narration of) are understood to be contingently linked to one another through contested political transitions in which different conceptions of industrial progress struggle for dominance. Conceptions of social identity are central to these conflicts and to their outcomes: The strategies of actors in struggles over the trajectory of industrial progress are shaped by images that they hold in their heads of the order of society, both as it is thought to be in the present and as it is desired to be in the future.

Most important in these conceptualizations of the social order are the way in which the position and rights of actors relative to others and relative to public institutions are conceived.

Transitional moments arise either through endogenously generated or exogenously induced problems in the social governance of economic activity, such as occurs in disputes over the development possibilities of a new technology (e.g., railroads in the 19th century or mass production and microelectronics in the 20th) and/or through confrontation with powerful foreign technological and organizational competition. Such moments give rise to alternative conceptions 9 of development and strategies for their realization. The struggles that ensue destabilize not only the institutional contours of the economy, but the structure and position of social groups and the way in which those groups understand their position in society and politics. The outcomes of such conflicts constitute, reconstruct, reposition and reconceive the entire range of social and political institutions that structure economic activity and, as such, shape the development of the division of labor, both within firms and within society as a whole.

There is a growing literature that adopts this kind of social constructionist understanding of the industrialization process. Most readers will be familiar with the arguments of this alternative tradition regarding small firms in the process of industrialization.22 In this literature, small firms are not coded in a teleological way as appropriate for a particular stage in the industrialization process and marginal in another. Rather, much of this alternative historical work has shown that decentralized, small and medium sized producer based industry not only existed as a viable alternative to the large industrial corporation in the past, but in certain places, such as Denmark and many parts of Germany, the political and economic conditions were actually established to sustain this kind of industrialization right into the present.23 In these accounts, the success or failure of the decentralized smaller producer based alternatives is attributed to the relative capacity on the part of the relevant actors to create institutions in the political economy that allow their conceptions of industrial progress to be realized.

The argument presented below applies this social constructionist framing of the politics of industrialization in general to the specific problem of construction of trade unions during the nineteenth century. The analysis regards trade unions as carriers not only of particular conceptions of what they want the social order in industrial society to be, but also of specific understandings of their own position, and the position of those they represent, in the existing social order. In other words, far from being adaptations to a given division of labor (or holdovers from some previous "stage" in the development of the division of labor), the aim of this analysis 10 will be to show that trade unions embody (or express or enact) particular understandings of what the social division of labor is (what Bourdieu calls the "social topology") and of how their members are positioned within it.

These broader characteristics of trade union identity may be seen clearly in the comparison of the emergence of craft unionism in the US and industrial unionism in Germany at the end of the 19th century. In both countries, conceptions of what trade union's were and who should belong to them competed with one another and changed over time. The final crystalizations were the products of immense social struggles during the mid and late 19th century about what the character and trajectory that the further development of the capitalistically structured artisan economy should be allowed to take. These struggles over the trajectory of industrialization, I claim, generated great disarray in the social structures of both societies. They created a range of new social actors, redefined relations between all social groups and threw into flux understandings that all actors had of their position within society--that is, of their social identities.

In the United States, the struggle was won by the advocates of the large corporation, mass production and progressive liberal order. Within this new social and political order, the craft union constituted a way to dignify and defend the status of the nativist, white, skilled craftsman.

In Germany, the same struggle over the character and trajectory of industrial development resulted in a set of complicated regional compromises that allowed a sophisticated decentralized industrial craft production to prosper along side large scale industrial corporations. In regions where the large corporation was dominant, trade unions were completely shut out, and in decentralized regions they competed with other artisanal organizations, not only for members, but over the definition of who was an industrial worker. In the end, a German industrial worker became one who, regardless of craft or skill level, belonged to no other organization in the regional labor market. By this means, membership in industrial unions gave such workers a coherent 11 identity and way of understanding and valuing their place in Wilhelmine society.

Early Nineteenth Century Artisinal Capitalism In Germany and the United States

The key first move in the development of this argument is to characterize the early 19th century artisan economies that existed in Germany and the United States prior to the giant struggle in the latter half of the century and lay out, at least the most salient, possible trajectories people wanted industrial development to take during the middle and late 19th century. The first step in doing this is to point out that in both countries the artisan economy in the early nineteenth century was a capitalist economy, and not an old honor bound, master centered one. In the "pre- capitalist" economy, there were, ideal typically, three actors: masters, journeymen and apprentices. Masters controlled the buying and selling of materials and there existed a wage structure that was established normatively (a "just price") and not by market forces.

Journeymen and apprentices, in the ideal model, were masters-in-waiting: They were expected to eventually attain the same social position as their masters (that is, by becoming masters themselves). The old artisinal economy, therefore, was not a class divided production economy, but a moral economy of craft honor, status and mutual obligation.24

During the first half of the nineteenth century, this "pre-capitalist" economy in both Germany and the United States broke down. Simplifying greatly, one could say that the old economic order was done in by the dramatic social and political changes that resulted in the pervasive diffusion of market exchange throughout both societies: Beginning already in the 18th century, product markets became more competitive and masters were forced to make changes in the character of production to keep pace: in particular they felt compelled to lower production costs by introducing market criteria into the setting of wages, or by dividing labor tasks in ways that made production more efficient or more flexible (or both). Some masters resorted to sub- contracting work out to cheaper, less skilled workers outside the workshop. Masters whose adjustment strategies met with success became entrepreneurs: they sought investment capital 12 from wealthy merchants and nascent bankers, expanded their production, sometimes extending the division of labor, trying to achieve economies of scale--etc. Unsuccessful masters either wound up increasingly dependent upon merchant putters-out, or they gave up their independence and sought work in shops (and factories) of more successful colleagues. This process of transformation introduced great ambiguity into the social identity of masters.25

At the same time, the economic pressures on masters totally disrupted the stable internal relations that had governed the artisan economy. The honorable world of the just price was converted to a world of market prices in which masters and journeymen calculated their relative capacity to acquire as much as they could of the surplus generated in production, based on the laws of supply and demand in the labor market. In this way, a common world was transformed into an adversarial one. Masters and journeymen (and unsuccessful masters) became capitalists and wage laborers. Early trade unionism emerged as a tactic by which journeymen might improve or protect their position in the labor market. Importantly, their battles with employers were always tinged with deep political outrage that the maneuverings of the capitalists constituted an attempt to undermine the rightful or just place of laboring craftsmen in society.26

Except for peculiar cases such as miners unions, these early journeymen's unions are characterized in most of the literature as craft unions.27 In many ways, however, the distinction between craft unions and industrial unions in this context is anachronistic, because the division of labor in production, despite the introduction of machinery, was still so rudimentary that the old craft basically constituted the new industry. Unions representing a craft in effect represented an industry.

In contrast to the standard view of industrialization outlined in the initial section of this note, the claim here is that it was not at all clear to the actors involved what direction the development of this early nineteenth century capitalist artisan economy would take. A variety of conceptions for development competed with one another, two of which we may, for the sake of simplicity, 13 consider here as most important. One envisioned a process resembling the traditional path of centralization, concentration, increasing scale, mass production, and so on. The other conceived of a more decentralized path in which flexible, small and medium sized producers of specialty products continuously absorbed new technologies and generated new products while utilizing and generating demand for skilled labor.28

The success of one or both of these trajectories in society presupposed, indeed would be determined by, the construction of a whole array of institutional arrangements regarding: the character of the financial system and the availability of investment capital; the organization of national transportation networks; regulations on the capacity of producers to cooperate in the market; the organization of industrial training; and the role of government in the economy.

Deciding about these things quite literally involved deciding about what kind of industrial society interested parties wanted to have: i.e., it puts in place the institutional framework that shapes economic activity and establishes what the distribution of power and status within the social and political community would be. Political struggle in postbellum America and in the Wilhelmine

Reich was in many respects dominated by these kinds of questions of industrial strategy, institutional design and social identity.

The details of these struggles and their arduous movement toward outcomes have been written about elsewhere and I will not deal with them here.29 Instead, I want to suggest how the outcome of this giant conflict in the 19th century affected the construction of trade union organizations in each country. Interestingly, in both countries, struggle over the definition of the trajectory of development gave rise to a similar split within the labor movement. In both cases, the split involved different conceptualizations of trade unions, of categories of workers and of the boundaries of the working class. These different conceptualizations of the social identity of trade unions in each case were made possible by the profound dislocation in the underlying topology of groups and relations in society that struggle over the constitution of an industrial 14 trajectory created.

On the one side, there were those who embraced the vision of a decentralized trajectory and advocated its institutionalization in a way that would transform the wage system into one of cooperative producers. The Knights of Labor in the United States and the Lassallean Socialists in Germany represented this tendency. Their associative vision of industrial society conceived not of a working class but of a more inclusive producing class. Both organizations, moreover, maintained that the creation and preservation of that class was consistent with the foundational social and political principles governing each society.30

On the other side were trade unionists and socialists influenced by the ideas of Karl Marx who believed that it was either utopian or politically impossible (or both) to push for the political transformation in basic property relations that their opponents in the labor movement advocated.

The labor movement, on this view, was too weak to influence the outcome of the struggle over the broad societal institutionalization of a particular trajectory of capitalist development and should instead focus on solidifying, defending and bolstering the rights and position of labor within whatever structure emerged. This conception of the trade union and its position in society was characteristic of both the pure-and-simple trade unionism of Samuel Gompers and the newly reconstituted industrial unionism in Wilhelmine Germany. It also describes the practice

(though not the ideology) of Wilhelmine social democracy.31

In both countries, the cooperative socialists lost and the marxian integrationists won. But here the similarities between the two cases ends. Setting aside the absence of a Social Democratic party in the United States, it remains the case that Gompers' pure-and-simple unionism was used to encourage the development of craft unions, whereas marxian trade unionism in Germany encouraged the development of industrial unions. To see how this came about, and to understand the crucial role of social identity in shaping the two outcomes, it will be necessary to look more 15 closely at each case.

Industrial Unionism in Germany and the Solidarity of Class

The key to understanding the formation of industrial unions in Germany is that the social position and identity of the skilled craftsman continued to be ambiguous in those regions where the industrialization of the artisan economy was ultimately institutionalized in a decentralized manner. Different kinds of organizations--industry associations, trade unions, craft chambers

(Innungen)-- competed for their allegiance and membership while they in turn sought membership or association with one of those organizations. For example, successful masters joined industry associations, unsuccessful ones craft chambers or trade unions. And, given the circulation and recombination of skills and assets that was characteristic of the highly specialized, decentralized production arrangements in these industrial districts, it was often the case that success and failure turned quickly into the other.32

Within this fluid socio-economic world, the conceptual burden on emergent organizations in their efforts to attract members was actually two-fold: they not only had to persuasively present themselves as representing positions in a social topology of groups and positions in the social structure, they had to construct a stable conception of what groups legitimately constituted or composed the social topology and for what reasons. For example, craft chambers, the representatives of the fin de siecle traditional Mittelstand, are widely interpreted as having had the burden not only of defining who they represented, but of how their members continued to have and deserve a respected place in the rapidly industrializing Wilhelmine Reich.33 The construction of organization, in other words, was intimately bound up with social and political power struggles over the redefinition of social identities and involved the invention of ideologies that elaborately characterized the growth and composition of modern society.34

How did this special two-fold burden affect trade union formation in a way that resulted in the formation of industrial unions? Here one must look to the particular way that decentralized 16 industrialization affected the structure of pre-capitalist guild membership in different industries.

As noted above, the traditional literature claims that labor unionism occurred independently of guild and journeymen organizations in the pre-capitalist artisinal economy. But this is only partly true. The historian Rudolf Boch has pointed out that the traditional argument only accounts for old master-centered guild organized crafts that were established in medieval or early modern times through grants from landed Aristocrats.35 These organizations, such as tailors, shoemakers, construction workers and many (though not all) workers in the metal trades, were structured in ways close to the ideal type outlined above, with the master completely in control of purchases and sales. When industrialization occurred in these artisinal trades, successful masters became capitalists, and in many cases transformed their guilds into industrial associations. Unsuccessful craftsmen and journeymen, on the other hand, were gradually cut loose from these older corporate affiliations and left to fend for themselves in the labor market and in the emerging new modern industrial society of status, citizenship rights and social hierarchy. We will deal with the fate of this later group in a moment, but the important point here is that the control and status of the master within the master centered guilds was transformed by the process of industrialization into the control and status of a capitalist employer.36

Boch points out, however, that this kind of dynamic was characteristic only of the master centered German guilds. There were other, differently structured guilds, not as ancient as the ones mentioned above and affected by industrialization differently. These alternative guilds were typically established during the eighteenth century, by city merchants who desired to organize a labor force to produce a particular article or commodity. In many cases the production arrangements established were organized on a decentralized, putting-out basis: Boch's best examples are the specialty cutting ware industries in Solingen and the specialty cotton and wool industries in the Kingdom of Saxony. But there were also cases in which work was performed in workshops, as in the printing trades.37 17

The distinctive feature of these newer guilds was the role of the master: he (always he's, even in putting out) did not control the purchases and sales of the trade as in the other guilds, but was simply a fully qualified craftsman. The position of control that the master occupied in the older guilds was in these newer guilds typically occupied by the merchant. Thus, when these trades began to industrialize, a clear cleavage emerged between merchants on the one side and the outworker guilds (masters, journeymen and apprentices) on the other. When guilds were legally outlawed, the outworkers organized themselves into what were called Fachvereine or associations of skilled outworkers. In the putting out industries cases, these Fachvereine were much like craft unions because they excluded workers with different craft skills who worked in different parts of the decentralized production process (e.g. forgers excluded grinders and vice versa in the cutlery trades).38 But, importantly, the continuity of membership between the

Fachvereine and the earlier merchant centered guilds left the status and social position of these workers in their communities less ambiguous than in those areas where, as we have seen, there was continuity with the guilds only among successful masters. The workers in the Fachvereine understood themselves, socially, economically and politically, to be a specific kind of independent, non-factory based industrial worker.

With the social status of the guild craftsmen transformed in the process of industrialization on the one hand into a set of entrepreneurial identities and on the other hand into a set of independent, non-factory worker identities, there remained a rather large social space left for workers who labored in factories and workshops. Such workers, as noted above, experienced the reality of their social position often as quite temporary: frequently they circulated through a variety of different positions, as factory worker, as independent sub-contractor, as small employer etc. But, given the way industrialization had transformed the social identity of the guild affiliations, these workers literally had no status and in many ways no sense of how to articulate their experience to themselves in social or political terms. 18

This kind of negative space in the social topology was ultimately given an identity by the social democratic movement and by allied trade unionists motivated by the marxian integrationist ideas discussed earlier. Both sets of actors constructed the idea of the working class and positioned it, conceptually, in a broader understanding of society and politics. They gave this working class, so defined, an historical mission and declared that it had (or deserved) social and political rights within German society. The ideas of the social democrats and their allied trade unionists made it possible for workers within the negative space in the social topology to recognize that they shared a social world. Most importantly, this connection reached across craft and skill distinctions: this common social space was not based on skill, but on the fact that all were factory workers cut off from old forms of social, corporate identity.

Inclusiveness marked the character of marxian influenced social democratic trade unions from their very beginning in the 1860s. To the late twentieth century eye these early unions appear to have been craft unions and many scholars, including Kocka, have described them as such. But this is once again only an indication of how inappropriate this distinction is for those early forms of artisanal capitalism. As elaborated earlier, because the internal division of labor within firms was so little developed, the craft was in many cases the industry. A better indication of the social and industrial inclusiveness of the marxian unions in the context of circulation of workers among different positions in the division of labor is that both workers and small employers belonged to the same unions: The strategy of the unions was to target those who had no other status or affiliation, not to focus on those who occupied the same position in the labor market.39

Seen from this perspective, the fact that the German trade unions opted for industrial and not craft unions after the anti-socialist law restrictions on labor union organizing were lifted in the 1890s makes sense. Even though industrialization of the artisan economy during the twelve years in which the unions were banned introduced new technologies and new skills into the 19 decentralized industries, the way in which the identity and status of the factory laborer had been constructed in German society by the marxism of the unions and the Social Democratic party made the formation of unions that organized differently skilled workers in the same factory, as well as their unskilled comrades, seem completely natural. The only dissenters to the formation of industrial unions were the independent non-factory workers in the Fachvereine, who had a completely different social identity.40

Craft Unions in the United States: Republicanism, Work, Respectability and Race

The social, institutional and political world in which artisanal capitalism in the United States was embedded was entirely different than in Germany. Social distinctions that were crucial in the construction of identity in Germany could not be made in the United States. There were, for example, no guilds in the US and as a result the crucial distinction between master-centered and merchant-centered guilds so important in creating an open space for the German marxism to construct the general skill spanning identity of a factory laborer could not be drawn here. Social identities were structured by different distinctions and these distinctions stemmed from the particular way in which artisinal capitalism in the United States was institutionalized and embedded in the surrounding society.

Seen comparatively, the key feature of artisinal capitalism in the United States was that the cleavages created between masters and journeymen by the breakdown of the traditional artisinal economy gave rise to the creation of "craft" unions of journeymen. This is important because journeymen were broadly recognized as having a social position and identity in society based upon their possession of a craft skill before the transition. By all accounts, these early journeymen unions used the relative stability of that social identity as a base upon which they could articulate a new "producer-republican" understanding of themselves, their union organizations and of the position and value of both within the early capitalist (Jacksonian) social order. The organizations and activities of journeymen craftsmen all presupposed this social 20 identity: the early "craft" unions that journeymen formed for purposes of strikes, their general hostility toward employers, their broad political activity-- all deployed and were embedded in a republican understanding of society and the polity. This understanding enabled the journeymen to articulate their grievances in terms of an offense to their dignity as skilled producers.41

Just as important as this producerist and republican transformation in the identity of journeymen craftsmen was the fact that the journeymen never carried the corporate, ascribed social and political status of guild members in Germany and other European countries. The formal political and legal status of the journeymen craftsmen in the United States was that of an individual citizen. This is important for a number of reasons, but most of all because it meant that the salient part of the journeymen's self conception and of the producer-republican ideology was not membership in a journeymen association or a craft union per se, but the possession of a skill.42

The fact that identity and social status came to be constructed as stemming from skill and not from corporate membership meant that anyone who possessed skill could make a claim to the status of producer and could understand themselves in those terms. This was true, naturally, of workers whose skills were generated through the process of industrialization, and of those who plied their trade in workshops and factories. This was not conceivable in the German case because the relationship between the possession of skill and corporate identity was far more complicated.

Now, if social status and identity within a producer-republican frame was linked to the possession of skill, it is easy to see that it would be important that others be able to recognize one who possessed it. The continuing salience of the linkage between skill and citizenship in the self understanding of workers and their organizations and especially the need for this kind of distinction, lays at the root of the strength of craft unionism in the United States throughout the period of nineteenth century industrialization. Workers in the United States understood their social position, their status, their respectability as citizens and their dignity as laboring beings as 21 rooted in their possession of skill and in the fact that possession of that skill was special and not common.

Throughout the nineteenth century and well into the twentieth, skilled workers and craft unions in the United States sought to protect their trades and to exclude from them all workers who would in some way diminish the social integrity of the trade. Crucially, because these were decisions about exclusion driven by social and political considerations of identity and not simply by calculations of market advantage, craft unions more frequently than not excluded not only unskilled workers, but also all workers who were coded in American society per se as undesirable and whose membership would therefore have undermined the social integrity of the trade. The classic "undesireables" for craft unions were black artisans and skilled immigrant laborers, neither of which was excluded on the basis of economic calculation or skill, but because they were either excluded outright or treated with great suspicion within American society generally.43

The paradox of American trade union development is that the evolution of this kind of exclusivist craft identity within the union movement was intensified by the historical transformation of the artisan economy into one dominated by large scale factory production. I propose, however, that the victory of what we now know as the modern large scale corporation was less important in determining this outcome than was the fact that more decentralized alternatives were defeated.

The advocates of these alternatives, in particular the Knights of Labor, self-consciously attempted to transform the producerist social identity within the laboring classes into a basis for more inclusive organizations and broader solidarity.

The idea of creating solidarity out of a conceptual understanding of society based upon distinction and difference was not as implausible at the time as it may at appear to us today. After all, the early antebellum producerist visions of a republican political economy were articulated by craftsmen laboring in that liminal zone referred to several times above in which 22 possession of a skill, membership in a craft and working in a particular industry were virtually overlapping activities. In that context, declaring yourself to be a producer and thus publically celebrating and defending the integrity of your skill, was not only a way to distinguish yourself from non-producers (the unskilled, bankers, lawyers, merchants etc), it also declared one's solidarity and commonality with other craftsmen. Expanding this latter dimension of the producerist understanding into a form of social identity that could have supported multi-skill industrial organization simply required a whole array of developments which never occurred, but which can be counterfactually posed (or simply reconstructed from what the advocates of more inclusive labor organizations actually said).

Consider the following. Suppose industrialization in the US actually had travelled along more decentralized craft industrial lines as it had in many regions of Germany. And, suppose further that the Knights of Labor were successful in institutionalizing that decentralization according to their associationalist principles (with producer cooperatives, an accommodating financial system and system of equitable national transport)44. Then the proliferation of new forms of skill and the circulation of producers through different positions in production that that kind of industrialization would have entailed might have sufficiently redefined the social understanding of who counted as a producer in a way that emphasized membership in a community of producers, rather than possession of skill.

The defeat of this alternative, more decentralized form of industrialization in the United States had the very important consequence of reenforcing the significance of skill as a basis for social status and social identity within the working class. This was true for two reasons.

First45, the dissolution of the Knights of Labor was precipitated by the notoriety of the Haymarket affair (regardless of how deserved the notoriety was), an event that tainted affiliation with the KOL with social disrespectability based on the manufactured impression that it was an 23 organization run by anarchists and foreigners.46 This unfavorable attribution caused a precipitous decline in membership in the KOL immediately following the Haymarket affair. But the scandal was also bad for the inclusive vision of the movement and for the understanding of an alternative ordering of society and politics that it represented. An essential aspect of the allure and strength of the KOL had been their ability to attach their controversial inclusivist and associational ideas to fundamental American values and myths: the movement continuously evoked, for example, the ideals of the American revolution in the articulation of its own program. The coding of the

Haymarket scandal in society at large made it increasingly impossible for the KOL to convincingly achieve this kind of ideological linkage, and left its inclusivist programs exposed charges of being un-American--and, by implication, incapable of manufacturing criteria for respectable citizenship. In the face of this kind of social risk and ambiguity, many white skilled workers removed themselves from the order and withdrew into a form of social organization that had already proven that it could preserve social respectability: the craft union.47

Secondly, this flight back to the political ideals associated with skill and the strategic orientation of the AFL craft unions which benefitted from the collapse of the KOL led, in turn, to the strengthening and institutionalization of craft division within the late nineteenth and early twentieth century American working class. Since the priority of the marxian/Gompersian trade unionists was to organize and defend the rights of workers in society as they found it and not to change the structure of the society in which those workers found themselves, they organized workers the way they wanted to be organized: that is, according to skill distinctions. Moreover, since the stated goal of these unions was to improve as best they could the position and integrity of laboring men in American society, they were extremely attentive to the way in which

American society itself distinguished a valued and respectable citizen from a suspicious or less desirous one.

The post 1890 history of the AFL is one of continuous capitulation to the reigning tenets of 24 organizational and social respectability. The central association, for example, capitulated to the repeated rulings of American courts that efforts to establish linkages across labor markets between skilled and unskilled violated fundamental norms underlying the constitutional order.48

Gompers, after a long and frustrating struggle, caved in to the intransigence of member unions and stopped pushing for racial inclusiveness among all AFL affiliates.49 On the eve of the United States' entry in the first world war, Gompers demonstrated the uprightness and thorough

Americanness of skilled, white, craft unionists by railing against the dangers of unrestricted immigration and lawless behavior on the part of immigrants.50

Conclusion:

The above account of the role of social identity in the construction of craft unions in the United

States and industrial unions in Germany before the first world war is admittedly far from definitive, even on its own terms. Much greater attention to actual discourses and debates in both countries, as well as greater elaboration of the way in which forms of social identity also involve relations of power in society. The aim of this note, however, has not been to make a definitive historical case, but to suggest and sketch out an alternative, social constructionist, way to think about the historical emergence of institutions in a political economy. Moreover, making an argument about the identity contours of organizational development is not inconsistent with arguments about their tactical and strategic behavior in the labor market or vis a vis the state. My point is simply that the self-definition of the trade union organizations involved ways of understanding and valuing the social world more broadly and that these forms of social understanding were crucial determinants for adoption of particular organizational forms over others.

There are two reasons that further development of the social constructionist perspective outlined here could be important in the context of current debates in historical and contemporary political economy both in Germany and the United States. First, traditional literatures in both countries 25 have been so driven by the cold war concerns about socialism and its presence or absence, that they have tended to overlook or take for granted very fundamental kinds of questions about the history of unions--such as what they meant and represented to their members. Very little of the literature on American exceptionalism, for example, makes explicit arguments about why there were craft unions in the United States. Their concern has been more with the political consciousness of workers and with the possibilities for the formation of a coherent left political movement in the United States. These are important questions, yet they reflect in many ways the range of the socially and politically possible that the particular constellation of debate during the cold war took for granted. In an age when liberalism and socialism, free markets and planning, no longer define the parameters of the social imagination, it is necessary to open up questions concerning the relationship between organizational form and social identities that the old dichotomies did not allow to be posed.

Second, concerns about the relationship between ideas of citizenship, social identity and the contours and role of labor market organization are very much central questions for the labor movement of today. The racial and ethnic contours of the labor market are changing dramatically both in the United States and Europe and the underlying structure of society, in a way analogous to the transformation in the 19th century, is being radically redefined by dramatic industrial, technological and political changes.51 Traditional structural and materialist approaches to trade unions too frequently fail to problematize the identity dimension of trade unions in our contemporary social and political world. Yet, unless one believes that material interests are not only transparent to individual actors, but that they suggest to the actors unique organizational forms for the pursuit of those interests, then the confrontation of questions of social identity and institutions as developed here in this research note cannot be avoided. 26

NOTES

1This research note was originally prepared for the conference "The Shifting Boundaries of Labor Politics: New Directions for Comparative Research and Theory" at the Center for European Studies, Harvard University, March 12-14, 1993. Some future version of it is likely to appear in an edited volume associated with that conference. I would like to thank the editors of this journal, particularly Karen Orren, and the following colleagues and friends for comments on earlier drafts: Michael Dawson, Vicky Hattam, Carol Horton, Richard Locke, Jeffrey Seitzer, George Steinmetz, Katherine Stone, Lowell Turner, Kim Voss, and the Wilder House workshops on Organization Theory and State-Building and on Comparative Politics and Historical Sociology at the University of Chicago.

2 This is a trope of much literature on industrialization, both in the United States and Europe. On the US case, two nicely paired works demonstrating the political range within the general view are Alfred Chandler, The Visible Hand, (Cambridge: Harvard/Belknap, 1977) and Martin Sklar, The Corporate Reconstruction of American Capitalism, 1890-1916. The Market, The Law and Politics, (New York: Cambridge, 1989). For Germany see the literature on "Organizierter Kapitalismus", e.g., H.A. Winkler, ed., Organisierter Kapitalismus. Voraussetzungen und Anfaenge, (Goettingen: Vandenhoeck & Ruprecht, 1972) as well as Juergen Kocka's extensive essay: "Entrepreneurs and Managers in German Industrialization" in Peter Mathias and MM Postan, eds., The Cambridge Economic History of Europe, Volume VII, Part one, (new York: Cambridge, 1978) pages 382-441

3a general kind of view of broad lines in trade union development and class formation is presented, with more subtlety than my account in the text gives it credit for, by Eric Hobsbawm, "The Making of the Working Class, 1870-1914" and "The `New Unionism' in Perspective" in Eric Hobsbawm in Workers: Worlds of Labor, (New York: Pantheon, 1984. Standard works in the historiography of American trade unionism, or the American labor movement more broadly conceived, adopt this kind of characterization of the historical evolution of trade union structure: see, e.g., David Brody, Workers in Industrial America: Essays in Twentieth Century Labor, (New York: Oxford University Press, 1980) or David Montgomery, The Fall of the House of Labor, (New York: Cambridge University Press, 1986). A representative German work is Klaus Schoenhoven, Expansion und Konzentration. Studien zur Entwicklung der Freien Gewerkschaften im Wilhelminischen Deutschland 1890-1914, (Stuttgart: Klett- Cotta, 1980). Juergen Kocka's "Problems of Working Class Formation in 27

Germany: The Early Years, 1800-1875" in Ira Katznelson and Aristide Zolberg, eds., Working Class Formation. Nineteenth-Century Patterns in Western Europe and the United States, (Princeton: Princeton University Press, 1986) represents the view developed in the text for the earlier period.

4for a methodological discussion of the range of explanations within such views, see Roberto Unger, Social Theory: Its Situation and Its Task, (New York: Cambridge University Press, 1987)

5generally on these newer views, see the essays in Peter Evans, Dietrich Rueschemeyer, and Theda Skocpol,eds., Bringing the State Back In, (New York: Cambridge University Press, 1985) and in Ira Katzenelson and Aristide Zolberg, eds., Working Class Formation. Nineteenth Century Patterns in Western Europe and the United States, (Princeton: Princeton University Press, 1986); On the United States, see, for example, Stephen Skowronek, Building a New American State, (New York: Cambridge University Press, 1982); William Forbath, Law and the Shaping of the American Labor Movement, (Cambridge, MA: Harvard University Press, 1991), and Leon Fink, Workingmen's Democracy: The Knights of Labor and American Politics, (Urbana: University of Illinois Press, 1983). For a Survey of this kind of literature in the German case, see Geoff Eley and David Blackbourn, The Peculiarities of German History, (New York: Oxford University Press, 1984) and John Breuilly, Labour and Liberalism in Nineteenth-Century Europe. Essays in Comparative History, (Manchester: Manchester University Press, 1992)

6see the rich account along these lines in Sean Wilentz, Chants Democratic. New York City and the Rise of the American Working Class, 1788-1850, (New York: Oxford, 1984) Chapters 1,3 and 6

7cf. David Montgomery, Workers Control in America, (New York: Cambridge, 1979) and Montgomery, The Fall of the House of Labor, op. cit. On the general transformation of the 19th century American industrial economy that Montgomery and others elaborate, see Alfred Chandler, Visible Hand, (Cambridge: Harvard/Belknap, 1977) and David Gordon, Richard Edwards and Michael Reich, Segmented Work, Divided Workers. The Historical Transformation of Labor in the United States, (New York: Cambridge, 1982)

8for an interesting contemporary account, very much in the frame being developed, see Theodore W. Glocker, "Amalgamation of Related Trades in American Unions" in American Economic Review, Volume 5, 1915, pages 554- 28

575

9John Dunlop, "The Changing Status of Labor" in Harold F. Williamson, ed., The Growth of the American Economy, (New York: Prentice-Hall, 1944) pages 607- 632; David Brody, "The Emergence of Mass Production Unionism" in idem., Workers in Industrial America. Essays in 20th Century Struggle, (New York: Oxford, 1980) pages 82-119

10Christiane Eisenberg, Deutsche und englische Gewerkschaften, (Goettingen: Vandenhoeck & Ruprecht, 1986)

11see Juergen Kocka's account of early trade union formation and the sources he cites there in "Problems of Working-Class Formation in Germany: The Early Years, 1800-1875" in Ira Katznelson and Aristide Zolberg, eds., Working Class Formation. Nineteenth-Century Patterns in Western Europe and the United States, (Princeton: Princeton University Press, 1986) pages 330ff

12ibid., page 338

13 The diffusion of the factory and large scale enterprise in the later half of the 19th century in Germany is outlined in Juergen Kocka's "Entrepreneurs and Managers in German Industrialization" in Peter Mathias and MM Postan, eds., The Cambridge Economic History of Europe, Volume VII, Part one, (New York: Cambridge, 1978) pages 382-441. Most accounts interpret the continued existence of small producers within a "dualist" framework in which small producers are understood to be backward, more fragile and less technologically sophisticated than producers in the large firm, or core, sector of the economy. The classic "dualist" statement on the fate of small producers in German industrialization is Wolfram Fischer, "Die Rolle des Kleingewerbes im wirtschaftlichen Wachstumsprozess in Deutschland 1850-1914" in Friedrich Luetge, ed., Wirtschaftliche und soziale Probleme der gewerblichen Entwicklung im 15.-16. und 19. Jahrhundert, (Stuttgart: Gustav Fischer Verlag, 1968) pages 131-143

14on the time during the anti-socialist law, see Gerhard A. Ritter and Klaus Tenfelde, "Der Durchbruch der Freien Gewerkschaften Deutschlands zur Massenbewegung im letzten Viertel des 19. Jahrhunderts" in Heinz Oskar Vetter, ed., Vom Sozialistengesetz zur Mitbestimmung, (Koeln: Bund Verlag, 1975) pages 61-120, esp. 69-88. On the organizational structure of the German "Free Trade Unions" after 1890, see Klaus Schoenhoven, Expansion und 29

Konzentration. Studien zur Entwicklung der Freien Gewerkschaften im Wilhelminischen Deutschalnd 1890 bis 1914, (Stuttgart: Klett-Cotta, 1980) esp pages 306-330.

15Schoenhoven, ibid. See also his "Localism--Craft Union--Industrial Union: Organizational Patterns in German Trade Unionism" in Wolfgang J. Mommsen and Hans-Gerhard Husung, eds., The Development of Trade Unionism in Great Britain and Germany, 1880-1914, London: George Allen and Unwin, 1985) pages 219- 238. On the strength of employers associations as an encouragement for industrial unions, see Schoenhoven, Die deutschen Gewerkschaften, (Frankfurt: Suhrkamp, 1987) pages 85-94 and Alexander Wende, Die Konzentrationsbewegung bei den deutschen Gewerkschaften, (Marburg: dissertation, 1913)

16For the Perlman argument, see Selig Perlman, A Theory of the Labor Movement, (New York: MacMillian, 1928) pages 182-200. For Ulman's subtle argument (one which seems to work much better for the IWW) see Lloyd Ulman, The Rise of the National Trade Union, (Cambridge MA: Harvard, 1955) pages 374-377

17specifically on the Knights collapse, see Kim Voss, "Disposition is Not Action: The Rise and Demise of the Knights of Labor" in Studies in American Political Development, 6(Fall 1992) pages 272-321. For a sprawling argument on the important role of employers in undermining American trade union efforts, in general, see Sanford Jacoby, "American Exceptionalism Revisited: The Importance of Management" in idem., ed., Masters to Managers. Historical and Comparative Perspectives on American employers, (New York: Columbia University Press, 1991) pages 173-200. For an argument about the shaping influence of the courts, see William Forbath, Law and the Shaping of the American Labor Movement, (Cambridge, MA: Harvard University Press, 1991), especially pages 68-79, 94-97.

18on the relative severity of German labor law on collective bargaining and union organization, see the early essays of Hugo Sinzheimer, in Hugo Sinzheimer, Arbeitsrecht und Rechtssoziologie. Gesammelte Aufsaetze und Reden, Band 1, (Frankfurt: Europaische Verlagsanstalt, 1976); Thilo Ramm, "The German Law of Collective Agreements" in Otto Kahn-Freund, ed., Labour Relations and the Law: A Comparative Study, (London: 1965) pages 84-91; Otto Kahn-Freund, Labour Law and Politics in the Weimar Republic, (Oxford: Basil Blackwell, 1981)

30

19see Schoenhoven, Expansion und Konzentration, op. cit., pages 91-149. The industrial structure of these regions of trade union strength (Saxony, the Rhineland, Wuerttemberg, Baden and Hesse) is outlined in chapter two of my book manuscript, Reconceptualizing The Sources of German Industrial Power, (Princeton Universtiy Press, forthcoming)

20Elizabeth Domansky-Davidsohn, "Der Grossbetrieb als Organisationsproblem des Deutschen Metallarbeiter-Verbandes vor dem Ersten Weltkrieg" in Hans Mommsen, ed., Arbeiterbewegung und industrieller Wandel. Studien zu gewerkschaftlichen Organisationsproblemen im Reich und an der Ruhr, (Wuppertal: Peter Hammer Verlag, 1980) pages 95-116

21This is the way that I understand the work of Roberto Unger, False Necessity Anti-Necessitarian Social Theory in the Service of Radical Democracy, (New York: Cambridge, 1988); Michael Piore and Charles F. Sabel, The Second Industrial Divide. Possibilities for Prosperity, (New York: Basic Books, 1984); and Charles Sabel Work and Politics (New York: Cambridge University Press, 1982). There are also strong affinities between the analysis I present below and the work of Pierre Bourdieu, in particular his "Social Space and the Genesis of `Classes'" in idem., Language and Symbolic Power, John Thompson, ed., (Cambridge, MA: Harvard University Press, 1991) pages 229-251, and more generally, Part Three of that same book, pages 163-251. Two other traditions also adopt the reversal suggested in the text, but make different kinds of claims about knowledge, causality and agency than the one's the above tradition advances. Here I have in mind, on the one hand, the "new institutionalism" in organization theory, eg.: Paul DiMaggio and Walter Powell, eds., The New Institutionalism in Organization Theory, (Chicago: University of Chicago Press, 1990) and George Thomas, John Meyer, Franscisco Ramierez and John Boli, Institutional Structure: Constituting State, Society and the Individual, (Newbury Park, CA: Sage, 1987). On the other hand there is post-structuralism, e.g. Michel Foucault, Discipline and Punish. The Birth of the Prison, (New York: Vintage, 1979)

22For general theoretical discussions of European and American industrialization in this way see the essays by Charles Sabel and Jonathan Zeitlin, "Historical Alternatives to Mass Production: Politics, Markets and Technology in Nineteenth- Century Industrialization" in Past and Present, 108 (August 1985) pages 133- 74 and Sabel and Zeitlin "Stories, Strategies and Structures: Rethinking Historical Alternatives to Mass Production" in idem and idem eds., Worlds of Possibility: Flexibility and Mass Production in Western Industrialization, (Volume 31

Prepared Under the Auspices of the Maison des Sciences de l'Homme, forthcoming)

23For a discussion of alternatives to the large scale corporation in the United States, see Gerald Berk, "Constituting Corporations and Markets: Railroads in Gilded Age Politics" in Studies in American Political Development, volume 4, 1990, pages 130-168 and Gary Herrigel, "Industry as a Form of Order: A Comparison of the Historical Development of the Machine Tool Industry in the United States and Germany" in Wolfgang Streeck, et. al., eds., Comparing Capitalist Economies: Variation in the Governance of Sectors, (New York: Oxford University Press, 1993). On decentralized industrialization in Germany, see my forthcoming book, Reconceptualizing the Sources of German Industrial Power, (Princeton University Press). On Denmark, see Charles Sabel and Peer Hull Christianson, "Denmark" in Sabel and Zeitlin, eds, Worlds of Possibility, ibid.

24My description of this kind of traditional economy is intended to be ideal typical. Actual pre-industrial artisan economies were far more complicated and had greater potential for conflict than my discussion in the text implies. A nice description of this kind of economy in 18th century America is presented in the first chapter of Sean Wilentz's Chants Democratic, (New York: Oxford, 1984). A case study of one such German artisan economy is Ekkard Wiest: Die Entwicklung des Nuernberger Gewerbes zwischen 1648-1806 [Stuttgart: Gustav Fischer Verlag,1968].

25On the transition from pre-capitalist to capitalist artisan economy in the United States, see David Gordon, Richard Edwards and Michael Reich, Segmented Work, Divided Workers, op. cit., pages 48-99; for the same transition in Germany, see my Reconceptualizing the Sources of German Industrial Power, op. cit., chapters 2 and 3.

26 Wilentz, op. cit., especially Chapters Six and Seven, makes this clear. Victoria Hattam's discussion of early trade union battles in the antebellum United States makes an instructive critique of Wilentz's conceptualization of the Journeymen's position in society and underscores the broadly political character of trade union rhetoric of the period, see Hattam, Labor Visions and State Power. The Origins of Business Unionism in the United States, (Princeton: Princeton University Press, 1993), pages 76-111. On early German trade unions, see Ulrich Engelhardt, `Nur vereinigt sind wir stark' Die Anfaenge der deutschen Gewerkschaftsbewegung 1862/63 bis 1869/70, (Stuttgart, 1977). See also the interesting essays by John Breuilly, Labour and Liberalism in nineteenth- 32

century Europe, (Manchester: Manchester University Press, 1992) especially "The Artisan economy, ideology and politics: the artisan contribution to the mid-nineteenth-century European labour movement" and "Liberalism or Social Democracy? Britain and Germany, 1850-1875" pages 76-159

27See again, Wilentz Chants Democratic, Chapter Six; and Kocka, pages 330ff

28This bifurcation in the debate is brought out clearly in Hattam's work, see Labor Visions and State Power, op. cit., pages 112-179. My own book, Reconceptualizing the Sources of German Industrial Power, op. cit., Chapters 2,3 and 4 develops an analogous range of alternatives in Germany.

29For my view of 19th century German industrialization, see Reconceptualizing The Sources of German Industrial Power, op. cit., chapters 2,3,&4. For the struggle between alternatives in the United States, in addition to Berk, "Constituting Corporations and Markets" op. cit., see Martin Sklar, The Corporate Reconstruction of American Capitalism, 1890-1916, (New York: Cambridge, 1989), Philip Scranton, Proprietary Capitalism, (Philadelphia: Temple, 1984), James Livingston, Origins of the Federal Reserve System. Money, Class and Corporate Capitalism, 1890-1913, (Ithaca: Cornell, 1986)

30See Vicky Hattam's discussion of the KOL from this point of view in Labor Visions and State Power, op. cit. A good overview of the KOL's general ideology is presented in the first two chapters of Leon Fink's Workingmen's Democracy, (Champaign-Urbana: University of Illinois Press, 1983). On Lassallean Socialism see William Harbutt Dawson, German Socialism and Ferdinand Lassalle, (London: Swan Sonnenschein & Co, 1891); Edward Bernstein, Ferdinand Lassalle as a Social Reformer, (London: Swan Sonnenschein, 1893); S. Na'aman, "Lassalle-- Demokratie und Sozialdemokratie" in Archiv fuer Sozialgeschichte,vol 3, 1963

31on Gomper's roots in marxian conceptions of the social role of trade unions in a developing capitalist society, see the excellent introduction to Gomper's autobiography by Nick Salvatore, "Introduction" in Samual Gompers, Seventy Years of Life and Labor. An Autobiography, (Ithaca: ILR Press, 1925) pages xi- xli, and also Stuart Bruce Kaufman, Samuel Gompers and the Origins of the American Federation of Labor, 1848-1896, (Westport CT: Greenwood Press, 1973) and H. M. Gitelman, "Adolph Strasser and the Origins of Pure and Simple Trade Unionism" in Daniel J. Leab, ed., The Labor History Reader, (Urbana: University of Illinois Press, 1985) pages 153-165. On the divide between revolutionary ideology and reformist practice in pre World War one German 33

social democracy one can still do little better than Carl E. Schorske, German Social Democracy, 1905-1917, (Cambridge MA: Harvard, 1955). Schorske is particularly good on importance of the trade unions in the, basically dominant, reformist faction . Mary Nolan's essay "Economic Crisis, State Policy and Working Class Formation in Germany, 1870-1900" in Ira Katznelson/ Aristide Zolberg, eds., Working Class Formation,(Princeton: Princeton University Press) pages 352-397 makes the point quite explicitly that the German Social Democrats did not pursue any policies in parliament that aimed at changing the basic institutional structures of the Reich in any way.

32for a discussion of this with examples, see chapter two of my manuscript The Sources of German Industrial Power

33An excellent discussion of the Mittelstand sensitive to this dimension is David Blackbourn, "The Mittelstand in German Society and Politics, 1871-1914" in Social History, number 4, January 1977, pages 409-433. The standard work on the politics of the lower middle classes (including the artisanate) in modern German history is Heinrich August Winkler, Liberalismus und Antiliberalismus. Studien zur politischen Sozialgeschichte des 19. und 20. Jahrhunderts, (Goettingen: Vandenhoeck & Ruprecht, 1979)

34cf. Bourdieu: "If the objective relations of power tend to reproduce themselves in visions of the social world which contribute to the permanence of those relations, this is therefore because the structuring principles of the world view are rooted in the objective structures of the social world and because the relations of power are also present in people's minds in the form of categories of perception of those relations. But the degree of indeterminacy and vagueness characteristic of the objects of the social world is, together with the practical, prereflexive and implicit character of the patterns of perception and evaluation which are applied to them, the Archimedean point which is objectively made available to truly political action. Knowledge of the social world and, more precisely, the categories which make it possible, are the stakes par excellence of the political struggle, a struggle which is inseparably theoretical and practical, over the power of preserving or transforming the social world by preserving or transforming the categories of perception of that world." in "Social Space and the Genesis of `Classes'" in idem, Language and Symbolic

Power, op. cit. 235-236 34

35Much of the following follows the analysis that Boch gives in "Zunfttradition und fruehe Gewerkschaftsbewegung. Ein Beitrag zu einer beginnenden Diskussion mit besonderer Beruecksichtigung des Handwerks im Verlagssystem" in Ulrich Wengenroth, Prekaere Selbstaendigkeit. Zur Standortbestimmung von Handwerk, Hausindustrie und Kleingewerbe im Industrialisierungsprozess, (Stuttgart: Franz Steiner Verlag, 1989) pages 37-70

36 Kocka treats this specific kind of evolution as the general one in his article "Problems of Working Class Formation in Germany" in Katznelson and Zolberg, eds., Working Class Formation, op. cit. pages 338-339. See also his general article on the emergence of entrepreneurs in German industrialization, Juergen Kocka, "Entrepreneurs and Managers in German Industrialization" in Peter Mathias and MM Postan, eds., The Cambridge Economic History of Europe, Vol. VII, (New York: Cambridge University Press, 1978

37cf. Boch, op. cit.

38on the cutlery workers in Solingen, see Boch's fabulous book, Handwerkersozialisten gegen Fabrikgesellschaft, (Goettingen: Vandenhoeck & Ruprecht, 1985). Boch's book is about the conflict that developed between the craft unions of the putting out tradesmen and the industrial unions of the factory cutlery workers during the first quarter of the twentieth century. The experience of the printing trades is a favorite example of most students of the labor movement interested in the creation of national unions because they were the first. See the discussion in Peter Swensen, Fair Shares, (Ithaca: Cornell University Press, 1989) and Gary Marks, Unions in Politics, (Princeton: Princeton University Press, 1989). The thing about the printers union is that it is always characterized as an industrial union, because it organized the entire industry. But in many ways there is alot of anachronism in that characterization for most of the nineteenth century because the organized craft was basically the organized industry. The distinction between craft union and industrial union is inappropriate for the case. See remarks above.

39see Kocka, "Problems of Working Class Formation in Germany" page 341.

40see Boch's rich discussion of the conflicts between the two kinds of worker identities in Solingen in his Handwerkersozialisten gegen Fabrikgesellschaft, op. cit. To avoid confusion it is important to note that though there were no large 35

conflicts over the idea of an industrial union in Germany beyond the conflict with the Fachvereine, there were many intra and inter-union conflicts during the 1890s and early twentieth century. In particular, there was tremendous debate over the creation of centralized national union structures. Localists were very strong during the 1890s. But, here again, even the localists had no problem with the formation industrial unions. See Schoenhoven, Expansion und Konzentration, op. cit.

41see the discussion of artisinal republicanism in Wilentz, Chants Democratic, op. cit.; and especially Hattam, Labor Visions and State Power, op. cit. Chapter 3

42Judith Sklar makes the connection between skill and citizenship in American Citizenship. The Quest for Inclusion, (Cambridge MA: Harvard University Press, 1991). This is also an important theme in Eric Foner, Free Soil, Free Labor, Free Men. The Ideology of the Republican Party before the Civil War, (New York: Oxford University Press, 1970)

43on the fate of black artisans and their exclusion from white trade unions, see W.E.B. Dubois, The Negro Artisan, (Atlanta: Atlanta University Press, 1902) which provides statistics on the progressive decline in their numbers after reconstruction. See also Harold M. Baron, "The Demand for Black Labor: Historical Notes on the Political Economy of Racism" in Radical America, vol. 5, No.2 (March-April, 1971). On the hostility of the AFL unions to immigrants, see Salvatore's introduction to Gompers, Seventy Five Years of Life and Labor, op. cit., and Gwendolyn Mink, Old Labor and New Immigrants in American Political Development: Union, party and State, 1875-1920, (Ithaca: Cornell University Press, 1986)

44 for a good outline of the producers vision of industrialization, see the discussion in Hattam, Labor Visions and State Power, op. cit. pages 122-129

45I thank Kim Voss for suggesting this general argument to me.

46Paul Averich, The Haymarket Tragedy, (Princeton: Princeton University Press, 1984), pages 215-240 and passim.

47see the interesting discussion of the early years of the machinist's union in Atlanta in Mark Perlman, The Machinists: A New Study in American Trade Unionism, (Cambridge: Harvard University Press, 1961). The craft union was founded by former Knights who felt that their reputations would be tarnished by 36

further associations with the order. The craft union they formed excluded all unskilled, all other crafts and all black workers.

48This is the message of Forbath, Law and the Shaping of the American Labor Movement, op. cit., though he doesn't appreciate the dimension of social identity that I am emphasizing. Hattam's, Labor Visions and State Power, op. cit. makes an argument that in some respects parallels Forbath's, but her argument is much more nuanced on identity questions.

49on racism and late nineteenth century AFL Unions, see Mark Karson and Ronald Radosh, "The American Federation of Labor and the Negro Worker, 1894-1949" in Julius Jacobson, ed., The Negro and the American Labor Movement, (New York: Anchor Books, 1968) pages 155-187. On the general increase of racism in late nineteenth and early twentieth century American political discourse, in all areas of social, political and economic life, see Carol Horton, Race, Liberalism and American Political Culture, Ph.D. dissertation (in progress), Department of Political Science, University of Chicago

50See Salvatore's Introduction to Gomper's autobiography, Seventy Years of Life and Labor, op. cit., pages xxxii-xxxvii as well as relevant sections of autobiography itself. See also Mink, Old Labor and New Immigrants in American Political Development, op. cit.

51For some who address problems of American trade unions from this perspective, see Michael Piore, "Post-Reaganomics: The Resurgence of the Social Sphere in Economic and Political Life? Part One" MIT Center for International Affairs, Working Paper Series, January 1989 and Alice Kessler Harris and Bertram Sliverman, "Beyond Industrial Unionism. Into Politics, Into Communities" in Dissent, Winter 1992, pages 61-66. English language literature on similar processes within European unions is very noticeably meager. One partial exception is Horst Kern and Charles Sabel,"Trade Unions and Decentralized Production: A Sketch of Strategic Problems in the West German Labor Movement" in Politics and Society, 19, no. 4 (1991), page 373-402